Proof-of-principle antimicrobial resistance routine diagnostics surveillance project (PoP project)

Transcription

1 PROTOCOL Proof-of-principle antimicrobial resistance routine diagnostics surveillance project (PoP project) Version 2.0

2

3 PROTOCOL Proof-of-principle antimicrobial resistance routine diagnostics surveillance project (PoP project) Version 2.0 Tjalling Leenstra Katherine Kooij Arjana Tambic Saskia Nahrgang Nienke van de Sande-Bruinsma

4 Abstract The Central Asian and Eastern European Surveillance of Antimicrobial Resistance Network (CAESAR) aims to address the lack of antimicrobial resistance (AMR) surveillance data in parts of the WHO European Region. Implementation of the Proof-of-Principle routine diagnostics project for antimicrobial resistance surveillance (PoP project) assists a country in building capacity to routinely perform antibiotic susceptibility testing at local hospital laboratories and at national reference laboratory level, by stimulating the utilization of blood culture diagnostics. This protocol provides the project coordination team with detailed information for the implementation of the project, including methods, standard operating procedures, and expected outputs of the project. As a result of the project, baseline data on the main pathogens causing blood stream infections and their antimicrobial susceptibility patterns will become available, providing clinicians with information needed to make antibiotic treatment choices. The collaboration between hospital laboratories and the national reference laboratory developed during the project lays down the foundation for a national AMR surveillance network. Keywords DRUG RESISTANCE, MICROBIAL ANTI-INFECTIVE AGENTS DIAGNOSTIC TESTS, ROUTINE - METHODS MICROBIOLOGICAL TECHNIQUES DATA COLLECTION GUIDELINE Suggested citation Leenstra T, Kooij K, Tambic A, Nahrgang S, van de Sande-Bruinsma N. Proof-of-principle antimicrobial resistance routine diagnostics surveillance project: Protocol. Copenhagen: WHO Regional Office for Europe; 2018 Address requests about publications of the WHO Regional Office for Europe to: Publications WHO Regional Office for Europe UN City, Marmorvej 51 DK-2100 Copenhagen Ø, Denmark Alternatively, complete an online request form for documentation, health information, or for permission to quote or translate, on the Regional Office website ( World Health Organization 2018 All rights reserved. The Regional Office for Europe of the World Health Organization welcomes requests for permission to reproduce or translate its publications, in part or in full. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers products does not imply that they are endorsed or recommended by the World Health Organization in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by the World Health Organization to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either express or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the World Health Organization be liable for damages arising from its use. The views expressed by authors, editors, or expert groups do not necessarily represent the decisions or the stated policy of the World Health Organization.

5 Contents Contributors iv Abbreviations v About this protocol vi Summary vii Introduction Project set up Goals and objectives Project design Coordination and technical expertise for the project Methodology Project procedures Data management and statistical analysis Quality assurance Safety considerations Follow-up Ethics Expected outputs of the project Problems anticipated Dissemination of Results References Annexes Annex 1. Assessment questionnaire for participation in the PoP project Annex 2. Terms of reference for a national coordination team for PoP project Annex 3. Clinical data form Annex 4. Laboratory results form Annex 5. Feedback form Annex 6. Evaluation form for weekly meetings Annex 7. SOP for sampling blood for culture Annex 8. SOP for processing samples Annex 9. Minimum set of antibiotics to be tested Annex 10. Patient information sheet Annex 11. Informed consent forms iii

6 Contributors Tjalling Leenstra MD PhD WHO Collaborating Centre for Antimicrobial Resistance Epidemiology and Surveillance, RIVM, Bilthoven, the Netherlands Katherine Kooij MD PhD WHO Collaborating Centre for Antimicrobial Resistance Epidemiology and Surveillance, RIVM, Bilthoven, the Netherlands Professor Arjana Tambic MD PhD University Hospital of Infectious Disease, Zagreb, Croatia Saskia Nahrgang MD MPH WHO Regional Office for Europe, Copenhagen, Denmark Nienke van de Sande-Bruinsma PhD WHO Regional Office for Europe, Copenhagen, Denmark iv

7 Abbreviations AMR AST BA BSI CA CAESAR CLSI EUCAST ID MAC PoP project SIRS SOP antimicrobial resistance antimicrobial susceptibility testing blood agar (plate) bloodstream infection chocolate agar (plate) Central Asian and Eastern European Surveillance of Antimicrobial Resistance Network Clinical and Laboratory Standards Institute European Committee on Antimicrobial Susceptibility Testing identification (of species) MacConkey (plate) Proof-of-principle antimicrobial resistance routine diagnostics surveillance project systemic inflammatory response syndrome standard operating procedure v

8 About this protocol This document is a generic protocol, containing guidance, recommendations and suggestions for the implementation of the Proof-of-Principle project. Based on continuous evaluation of ongoing and past PoP projects, adaptations and improvements are made to the generic protocol 1. The protocol provides guidance on active case finding, aseptic venipuncture procedures, and laboratory procedures, including timely feedback of results. In addition, it provides guidance on species identification and the use of EUCAST for antimicrobial susceptibility testing. The minimum set of antibiotics to be tested is based on recommendations from the CAESAR manual (1). It may be beneficial to adapt the protocol to fit the financial and organizational context, and to better embed the PoP protocol within locally used protocols and guidelines. All adaptations should be implemented in collaboration with the WHO Country Office and WHO Regional Office for Europe. Detailed SOPs and sample forms are included as Annexes. It may be beneficial to adapt these prior to implementation of the protocol. Ethical approval of the project should be obtained at country level prior to implementation of the protocol. 1 Previous versions of the protocol have not been published online vi

9 Summary The WHO Global Action Plan on antimicrobial resistance (AMR) urges countries to develop a national surveillance system for antimicrobial resistance to strengthen their knowledge about AMR. The recent WHO global report on surveillance of antimicrobial resistance revealed that large gaps in information on antimicrobial resistance exist because many countries do not have a national surveillance system in place. An important limiting factor to conducting routine AMR surveillance in a country is the underutilization of bacteriological diagnostics in routine clinical practice. A recent assessment led by the WHO Regional Office for Europe identified the following constraints for the use of bacteriological diagnostics in routine clinical practice: (i) financial constraints such as lack of reimbursement for diagnostic testing, (ii) limited perceived benefit for performing such diagnostic tests by clinicians, and (iii) weak communication and linkages between hospital laboratories and clinical wards. This project wants to demonstrate that use of routine bacteriological diagnostics to guide treatment decisions, and AMR surveillance provides essential information to care providers at the point of care. At the same time it can help to showcase how surveillance plays a key role in targeting infection prevention and control policies in clinical settings throughout the country. More specifically, this project (i) supports the collection of blood cultures from patients with a clinical suspicion for bloodstream infection (BSI) and (ii) facilitates bacteriological processing of microbiological samples, including species identification (ID) and antimicrobial susceptibility testing (AST). A limited number of eligible hospitals, typically between three and five implement the project protocol, and are accompanied in that by a team of trained project coordinators, who provide follow-up and guidance throughout the process. In addition to technical guidance, material support is provided to clinicians and hospital laboratories, based on a needs assessment and process mapping before the start of project implementation. Samples collected as part of the project are processed twice in parallel at the hospital laboratory and at the national reference laboratory for AMR to (i) ensure process quality and (ii) to provide a training opportunity for the laboratories involved. As a result this project generates baseline data on the main pathogens causing BSIs and their antimicrobial susceptibility patterns. In the process of generating these data, the capacity for diagnostic bacteriological testing at the local hospital laboratory and at the national reference laboratory is strengthened. With a view on creating sustainable practice, the project also supports the creation of a sustainable structure for AMR data sharing and bacterial isolates collection, which builds the foundation for a national AMR surveillance network. vii

10 viii

11 Introduction AMR is a threat to adequate treatment of infectious diseases in individual patients and has been associated with higher morbidity, higher mortality and longer hospital stays; consequently it has effects on both the individual and on wider society (2). Since resistance levels are currently rising worldwide (3 5), there is a need for better evidence to support health policy decision-making. Local, national and international surveillance programmes that monitor the magnitude and trends in AMR can support the development of appropriate clinical guidelines, support empirical antibiotic therapy and guide infection control interventions. Information derived from surveillance also plays an important role in creating awareness for AMR among health care professionals, policy-makers and the general public (6). The Central Asian and Eastern European Surveillance of Antimicrobial Resistance Network (CAESAR) aims to address the lack of surveillance data in parts of the WHO European Region by building capacity for surveillance, strengthening laboratory capacity and quality, and developing a network of national AMR surveillance systems for countries in the Region. CAESAR started in October 2012 as a joint initiative of the WHO Regional Office for Europe, the European Society of Clinical Microbiology and Infectious Diseases and the Netherlands National Institute for Public Health and the Environment (1). By initiating collaboration with CAESAR, a country can identify a series of steps necessary to address the issue of AMR, including setting up a functional national AMR surveillance system. Underutilization of microbiological diagnostics, resulting in a low number of blood cultures being processed, tested for presence of bacterial pathogens and investigated for antimicrobial susceptibility or resistance, has been identified as an important obstacle to performing national AMR surveillance. This underuse of clinical diagnostics may furthermore have a negative effect on data quality as it leads to an overrepresentation of critically ill patients or patients in whom first-line antibiotic therapy did not work and it can bias surveillance results towards higher resistance numbers (2,7). Most initial antibiotic prescribing is empirically based on the clinician s decision that a patient is likely to have a bacterial infection and on his or her knowledge of the most likely pathogens and their likely antibiotic susceptibility. Empirical choices must often be broad spectrum to cover many possible pathogens; however this approach increases the selective pressure for resistance. AMR threatens to undermine the efficacy of empirical choices, which results in poorer outcomes for individual patients. Appropriate use of bacteriological diagnostics should support clinical prescribing decisions for the individual patient, and provide information about the causative pathogen, and its pattern of susceptibility or resistance to antibiotics. The last can be used to guide therapy, confirming in some cases that the empirical choice was appropriate; identifying patients for whom the initial broad-spectrum therapy can be de-escalated while still providing adequate coverage; and identifying patients for whom the empirical choice was inappropriate because the infecting organism was resistant. Aggregating diagnostic data obtained over time for many individuals can provide an important source of surveillance information, which increases knowledge of pathogens dominating locally and their likeliest drug resistances; it can thus shape empirical prescribing policies for the benefit of future patients. The objective of this project is identifying bacteria causing BSIs and their antibiotic susceptibility patterns. A BSI is defined as the presence of bacteria in the blood as demonstrated by blood culture. This is taken to include all clinical manifestations of BSIs, including sepsis (life-threatening organ dysfunction caused by a dysregulated host response to infection) and septic shock (sepsis with persistent hypotension despite fluid resuscitation plus cellular/metabolic abnormalities) (8). The systemic inflammatory response syndrome (SIRS) criteria proposed by the American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference Committee (9) will be used as a means of early detection of patients with a BSI 1

12 and to indicate that a blood culture should be considered. Additional diagnostic criteria are required for children (10) and for identifying patients with sepsis, severe sepsis and septic shock (11). The proposed project aims to provide insight into the current levels of resistance in a country by stimulating the appropriate collection of clinical samples (blood samples) at the hospital level and by routine surveillance at local and central laboratories of microorganisms causing BSIs. In the process, clinicians will receive timely feedback from microbiology laboratories as part of the diagnostic work-up of patients with infectious diseases, including those with hospital-acquired infections. It is expected that this feedback of individual results, together with the reports of average resistance in the whole population (cumulative antibiograms), will enhance the perceived utility of performing microbiological testing to guide treatment decisions among clinicians. Furthermore, the higher number of samples to be processed at the hospital and national reference laboratory levels will provide practical learning opportunities for microbiologists at both locations. 2

13 Project set-up Goals and objectives Main goal: The main overarching goal of the project is to contribute to improving clinical care for patients admitted with suspected BSIs, in accordance with evidence-based medicine. Secondary goals of the project are: to demonstrate to clinicians the value of clinical microbiology as part of the diagnostic work-up of patients with suspected BSIs and to improve the clinical work-up and timely feedback of laboratory results to prescribers of antimicrobial drugs, thus allowing for optimization of antimicrobial therapy; to establish and support a surveillance network as a starting point for a functional national sentinel laboratory-based surveillance system for AMR; to provide insight into antibiotic resistance levels in a country. The projects objectives are: Ad 1. improve active case finding practice; improve turnaround time at key steps of blood sample processing at the local laboratories; improve concordance between local and reference laboratory in reaching accurate test results; ensure multidisciplinary interaction around a patient. Ad 2. strengthen the AMR reference and surveillance capacity at the national reference laboratory, by providing: - terms of reference; - training on international guidelines (EUCAST, laboratory quality stepwise implementation (LQSI); - training on how to train local laboratory personnel; - support to set up national external quality assessment scheme (NEQAS); strengthen capacity at the local laboratories, by providing - training on international guidelines (EUCAST, LQSI); ensure transport of samples, isolates, and data between local laboratories, reference laboratory and national AMR surveillance group. Ad 2 and 3. increase the volume of clinical samples processed by local and reference laboratories, thus building laboratory capacity. 3

14 Ad 3. contribute local AMR data to a national database; improve the capacity to manage, analyse, interpret, and report on AMR data, by providing training; provide a top 5 ranking of organisms and a cumulative antibiogram; provide data for national and regional surveillance. Project design Type of project This project is set up as a descriptive observational study. Research population All consecutive patients with suspected BSI at time of admission or when already admitted to hospital during the course of the project will be asked to participate. Sampling frame Three to five hospitals routinely treating patients with BSIs will be selected to participate in the project. Recruitment through active case finding will take place among patients admitted to hospital departments that take patients with suspected BSIs from the community (e.g. emergency departments) and wards where patients are at risk of developing hospital-acquired BSIs (e.g. intensive care, urology or surgical departments). The hospitals participating in the project will be selected using an assessment questionnaire for participation in the PoP project (Annex 1). As a prerequisite, the hospitals need to routinely provide treatment to patients with BSIs, and have either on-site bacteriological laboratory diagnostic capacity (species ID and AST) or an agreement with an external laboratory capable of providing bacteriological diagnostic services (species ID and AST). Specific hospital departments or wards will be targeted to focus on active case finding and for capacity-building/technical support. Active case finding and inclusion of patients Clinicians (medical doctors and nurses) will be trained to conduct active case finding (see Quality assurance). Sequential patients with clinical suspicion of BSI or sepsis will be included in the project. The SIRS criteria will be used for early recognition of patients with suspected BSI or sepsis (9). The SIRS criteria are sensitive criteria aimed at identifying as many patients with a potential BSI as possible (11). For adults, the SIRS criteria are two or more ( 2) of the following (9) core body temperature >38.3 C (fever) or <36 C (hypothermia) heart rate >90 beats/min (tachycardia) respiratory rate >20 breaths/min (tachypnoea) leukocytes > /l or < /l or >10% immature neutrophils (band forms). Age-appropriate criteria will be used for children as proposed by the International Consensus Conference on Paediatric Sepsis (10). For consistency with local practice, the age-specific criteria for tachycardia and tachypnoea used may be adapted to local guidelines. For children, the SIRS criteria are two or more ( 2) of the following (10): 4

15 core body temperature >38.3 C (fever) or <36 C (hypothermia) tachycardia defined as a mean heart rate more than two standard deviations above normal for age (age-specific criteria; Table 1) or for children younger than 1 year of age, bradycardia defined as a mean heart rate <10th percentile for age tachypnoea defined as a respiratory rate more than two standard deviations above the normal for age (age-specific criteria; Table 1) or mechanical ventilation for an acute pulmonary process elevated or depressed leukocyte count (according to age-specific criteria; Table 1) or >10% immature neutrophils (band forms). Table 1. Age-specific criteria for SIRS Age group Core body temperature ( C) Heart rate (beats/min) Fever Hypothermia Tachycardia Bradycardia Respiratory rate (breaths/ min) Leukocyte count (10 9 /l) Newborn (0 days to 1 week) >38.3 <36.0 > >50 >34 Neonate (1 week to 1 month) >38.3 <36.0 >180 <100 >40 >19.5 or <5 Infant (1 month to 1 year) >38.3 <36.0 >180 <90 >34 >17.5 or <5 Toddler and preschool (>1 to 5 years) >38.3 <36.0 >140 NA >22 >15.5 or <6 School age (>5 to 12 years) >38.3 <36.0 >130 NA >18 >13.5 or <4.5 Adolescent (>12 to 18 years) >38.3 <36.0 >110 NA >14 >11 or <4.5 Adult (>18 years) >38.3 <36.0 >90 NA >20 >12 or <4 Source: Bone et al (9) and Goldstein et al (10). All patients meeting these SIRS criteria will be clinically evaluated by a medical doctor. If the doctor s clinical judgement is that the patient may have a BSI or sepsis (any stage), blood for culturing will be taken prior to initiation of antibiotic therapy. For clinical diagnostic criteria for suspected BSI or sepsis, prevailing clinical guidelines (12,13) or WHO guidelines (14,15) are used. A practical approach could be to take a sample for blood culture from all patients with a clinical suspicion of a serious bacterial infection in whom (intravenous) antibiotic therapy is considered. The clinical utility of the SIRS criteria for detecting patients with sepsis has recently been called into question because of its low specificity and imperfect sensitivity (8). A recent large-scale retrospective study in the United States has identified alternative clinical criteria (altered mentation, low systolic blood pressure and elevated respiratory rate) to identify adults with sepsis (16). However, these criteria have not yet been validated prospectively in a wide range of clinical settings, including low-resource settings, and were not adopted in the 2016 international guidelines for clinical management of sepsis (17). Until 5

16 there is a clear consensus about an alternative, this project will use SIRS for early detection of patients with suspected BSI in whom a blood culture should be considered. Exclusion criteria Patients not admitted to one of the selected wards Patients not consenting to inclusion. Withdrawal criteria Patients will only be subject to project procedures once they have agreed to inclusion. Information about patients following the initial blood sample will be secondary on collection of clinical information from the treating physician. A patient can withdraw from the project at any time. Sample size calculation Having an adequate sample size is important to ensure the objectives of a study are attained. The main endpoint for estimating required sample size in this project uses the weighted average susceptibility for any one of the antibiotic choices for (empirical) treatment of a BSI caused by a Gram-negative bacterium. The weighted average is calculated as shown in Table 2. Importantly, the denominator for calculating the weighted average is the total number of relevant microorganisms isolated. Based on data collected by CAESAR in 2015, a rough expectation of rank-ordering and proportions non-susceptibility is made; approximately two thirds of all isolates are expected to be Gram negative. Table 2. Calculation of weighted average antibiotic susceptibility for BSI caused by Gram-negative bacteria Proportion (%) No. a NS (%) b Absolute NS c Weighted average NS Escherichia coli % 32 Klebsiella pneumoniae % 30 Pseudomonas aeruginosa % 6 Acinetobacter spp IR 40 Total % (108/200) Notes: NS: Isolates non-susceptible (intermediate or resistant) to third-generation cephalosporins; IR: Intrinsic resistance (assumed that 100% of the species are NS); a Total number of isolates for a given bacterial species assuming a total sample size of 200. b Percentage of isolates not susceptible to third-generation cephalosporins. c Total number of isolates not susceptible to third-generation cephalosporins. Source: Based on data collected by CAESAR in 2015 (1). Sample sizes in Fig. 1 are estimated using the formula for one-sample comparison of a proportion to a prespecified proportion (18). More specifically, we want to assess whether susceptibility to a particular antibiotic or combination of antibiotics is significantly higher than 10%, which is often used as the critical threshold of non-susceptibility for which a certain treatment option is deemed not fit for empirical treatment of patients with life-threatening disease. 6

17 Fig. 1. Required sample sizes for finding a statistically significant difference between observed proportion and a critical threshold of 10% based on a one-sample test for comparison of proportions for given confidence level and power Required sample size (95% confidence interval) Observed proportion non-susceptible (%) 80% power 90% power Notes: Confidence interval is 100% minus the chance of a type I error (false-positive finding), or 100% minus the chance of incorrectly concluding the observed proportion is different from 5%; Power is 1 minus the type II error (false-negative finding) or 1 minus the chance of incorrectly concluding there is no difference between the observed proportion and 10%. A total sample of 200 Gram-negative blood culture isolates (one per episode per patient) would be sufficient that an observed proportion of non-susceptibility of 16.5% would be statistically different from the threshold proportion of non-susceptibility of 10% (80% power and 95% confidence interval). Assuming that two thirds of bloodstream isolates are Gram negative and a 15% positivity rate of blood cultures, a total of 2000 blood cultures would need to be taken and processed. Expected duration of the project The expected duration of the project is six months. After the six months, the attained sample size will be assessed and a decision will be made either to stop or to continue the project until the required sample size has been attained (Table 3). Table 3. Planned project duration Preparations Year 1 Year 2 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 J F M A M J J A S O N D J F M A M J J A S O N D Training Start Analysis Review 7

18 Coordination and technical expertise for the project A country project will require the following participation: Project initiator WHO Collaborating Centre for Antimicrobial Resistance Epidemiology and Surveillance, National Institute for Public Health and the Environment, the Netherlands. Principal investigator: Tjalling Leenstra Technical support team WHO Collaborative Centre for Antimicrobial Resistance Epidemiology and Surveillance, National Institute for Public Health and the Environment, the Netherlands Department of Clinical Microbiology, University Hospital of Infectious Disease, Zagreb, Croatia Programme Control of Antimicrobial Resistance, WHO Regional Office for Europe, Copenhagen, Denmark WHO country office Project coordination team National AMR focal point Coordinator and data manager Research coordinator and quality assurance specialist Epidemiologist bacteriologist Clinician Support personnel (laboratory technician) Support personnel (communication, public health/ministry of Health) Support personnel (procurement support, short term) Terms of references for a national coordination team for the PoP project are listed in Annex 2. Project sites Three to five research sites with each providing: clinical epidemiologist clinical bacteriologist clinician Clinical laboratories Three to five linked laboratories will be involved in the project. Reference laboratory A national reference laboratory for checking local results and creating a database of the information generated. 8

19 Methodology Project procedures The project procedures (work flow) are depicted in Fig. 2. The dotted lines reflect feedback of laboratory test results from the local laboratory to the clinic for updating patient therapy and from the reference laboratory to the local laboratory for quality assurance. Throughout the duration of the project, data are collected to monitor the progress and evaluate the results of the project. The following data collection forms are included in the Annex: a clinical data form (Annex 3), a laboratory results form (Annex 4), a feedback form (Annex 5), and an evaluation form for weekly meetings (Annex 6). Active case finding All clinicians, including nursing staff, working on selected departments/wards will be asked to support the project and the collection of samples for blood culturing. All clinicians will be trained in recognizing patients fulfilling the SIRS criteria. This approach will raise awareness about the project being done and maximize inclusion of patients meeting inclusion criteria. Patients should be included whenever they develop signs of a BSI or sepsis. All patients will be actively screened at least twice a day to increase timely recognition and inclusion of patients. Blood sampling Annex 7 outlines the SOP for blood sampling procedures. Clinicians taking the blood culture samples from patients will receive training in proper sampling techniques at the start of the project and will be provided with SOPs. Samples for blood culture should be taken from all patients with suspected BSI, after evaluation and confirmation of the indication for blood culture by a physician. Fig. 2. Project procedures REFERENCE LAB Confirming testing CLINIC Active case finding SIRS criteria CLINIC Blood sampling LABORATORY Blood culturing LABORATORY Preleminary report (Gram stain) Quality assurance LABORATORY Final report Species ID CLINIC AST Update therapy if required 9

20 Since antibiotic therapy may lead to selection of resistant pathogens, blood samples should, whenever possible, be taken prior to starting antibiotic therapy. However, because resistance proportions should reflect resistance levels at admission to the hospital and during hospital stay, antibiotic therapy initiated prior to admission to hospital will not be a reason for exclusion. Additionally, patients already admitted who are a suspected of having a BSI developing during antibiotic treatment or who have a recurrence of symptoms after a period of remission or quiescence following initial antibiotic treatment will not be excluded. Distinction between patients with or without prior treatment will be made using the information collected on the clinical data form (Annex 3). Taking blood for culture should not delay the initiation of antibiotic therapy. To ensure this, sampling materials should be available at all selected departments/wards. Puncture sites will be carefully disinfected; the site is cleaned with alcohol followed by chlorhexidine and allowed to dry. Note. If possible, sampling from indwelling catheters should be avoided. Tops for blood culture bottles will be disinfected with 70% isopropyl alcohol (propan-2-ol; alcohol pad). For adults, two blood sample sets will be taken by venipuncture from separate sites with a maximum of 30 minutes apart, provided that the situation of the patient allows this. Each set contains two blood culture bottles of 10 ml blood each. For paediatric patients, the blood volumes taken vary with patient weight (see Annex 7 for details). The blood withdrawn should be immediately inoculated directly into the appropriate blood culture vial(s) without changing the needle on the syringe. Sample processing at the laboratory Annex 8 outlines the detailed sample processing steps at the laboratory. After sampling, the blood cultures are stored at the hospital laboratory (or contracted external laboratory) for incubation for a maximum of seven days. Blind subcultures will be performed after 24 h, 48 h, and seven days incubation at the site. If bacterial growth is detected, species ID and AST will be conducted on site. For reasons of quality assurance, each positive blood culture isolate will be sent to the national reference laboratory for parallel ID and AST. In addition, reference laboratory will perform confirmation of any special resistance patterns and further characterization of resistance mechanisms. Blood culturing Blood culturing procedures are described in detail in the SOP for sample processing (Annex 8). Number of cultures The optimal number of blood cultures that should be obtained varies with the clinical condition, suspicion for underlying infection (e.g. the pretest probability), and the urgency of the need for treatment (19). The total volume of blood drawn may be more important than the number of sets taken at the same time. In this study by Patel et al, collection of two aerobic and one anaerobic blood culture bottles per set resulted in improved yield compared to two bottles per set (10 ml per bottle) (20). Among 134 patients in the same study, the yield from two blood culture sets (30 ml in three bottles per set) was the same as the yield from three blood culture sets (20 ml in two bottles per set) (20). Another study compared the yield of various numbers of blood cultures taken within 24 hours (20 ml in two bottles per culture set). The cumulative yield of true pathogens increased from 65% in the first culture to 80% including the second culture, 96% including the third culture, and 100% including the fourth culture (21). 10

21 Species ID Standard methods for species ID will be used. All isolates should be identified to genus level. The following bacterial species should be fully identified: Streptococcus pneumoniae (STRPNE) Staphylococcus aureus (STAAUR) Enterococcus faecalis (ENCFAE) Enterococcus faecium (ENCFAI) Escherichia coli (ESCCOL) Klebsiella pneumoniae (KLEPNE) Pseudomonas aeruginosa (PSEAER) Acinetobacter spp. Nota Bene: Coagulase-negative staphylococci should be identified as such. Other isolates should also be recorded but could be identified to genus level. AST AST will be done by disc diffusion according to European Committee on Antimicrobial Susceptibility Testing (EUCAST) standards (22). Annex 9 contains a minimum set of antibiotics to be tested per microorganism under surveillance. These bug drug combinations are based on the recommendations in the CAESAR manual (1), including indicator antibiotics for the main antibiotic groups plus some empirical treatment options not in the CAESAR manual. Isolate storage and shipment Shipping and transportation of isolates will be performed according to the recommendation provided in SOPs of the national AMR reference laboratory. Positive isolates and accompanying clinical data forms will be collected on the same day by a driver from the national AMR reference laboratory. The local team (microbiologist) will contact the coordinator team (microbiologist) about the existence of a positive sample and the national AMR reference laboratory will organize transport in timely manner. All isolates cultured at the local laboratories will be shipped to the national AMR reference laboratory for duplicate testing at central level. Feedback of laboratory results Preliminary and final laboratory results should be noted on the laboratory results from see Annex 4. Each clinical laboratory should appoint a (clinical) microbiologist responsible for providing feedback to clinicians on preliminary results (Gram stain of a positive blood culture) and final results of blood cultures (species ID and AST). Feedback should be timely and frequent (as soon as results become available) to establish a strong working relationship. It is expected that most results will be final within 72 hours. It is recommended that the microbiologist provides preliminary results to the clinician over the telephone. The local laboratory is responsible for informing the clinicians on species ID and AST results based on local testing without waiting for confirmation by national AMR reference laboratory. The process of communication and feedback of results between the microbiologist and clinician will be registered on the feedback form (Annex 5). 11

22 Weekly meeting A weekly meeting with clinicians, clinical epidemiologists and the clinical microbiologist will be held to discuss patients and the laboratory results and any implications for antibiotic treatment at each of the project sites. The project coordination team will be present at this meeting and will fill out an evaluation form to monitor progress of the project (Annex 6). In addition, timely and frequent feedback on results of confirmatory testing done at the national reference laboratory should be provided to hospital laboratories to allow timely trouble-shooting and adjustment of procedures in case of discrepant results. Roles and responsibilities of the teams at the project sites Clinical microbiologist: sample processing, blood culturing, species ID and AST; feedback of laboratory results to clinician; filling in laboratory forms; participating in weekly meetings Clinicians: active case finding, obtaining consent for patient participation (signed informed consent form), blood sampling, incorporating laboratory results for treatment adjustments, filling in clinical forms, participating in weekly meetings Clinical epidemiologists: active case finding, coordination of data collection and completion of forms, participating in weekly meetings. Data management and statistical analysis Unique patient identifier The unique patient identifier (patient registration number) used in the hospital together with the date of sampling will be used for the unique identifier combination for the patient. Both are required for merging of the data from different sources. This combination should be present on all forms and patient materials. Laboratory data Laboratory results will be entered on paper laboratory result forms (Annex 4) and isolate record forms. Copies of the laboratory results and isolate record forms will be collected by the study coordinator on a weekly basis and checked for completeness and consistency. Whenever necessary, data are completed and/or corrected in consultation with the hospital laboratory. Data will then be entered/compiled in a dedicated database. Data will be entered twice using predesigned data entry screens (23) and any discrepancies manually corrected. Clinical data With each clinical episode (i.e. every time blood cultures are taken), the clinical data form should be filled in (Annex 3). Copies of the clinical data forms will be collected by the project coordinator on a weekly basis and checked for completeness and consistency. Where necessary, forms will be completed and/or corrected after consultation with the clinician. Data will then be entered into a dedicated database. Data will be entered twice using predesigned data entry screens (23) and any discrepancies manually corrected. Confidentiality of patient data All data and diagnostic specimens transferred to the project coordinator and the reference laboratory will be devoid of any personal identifying information (name, date of birth, residential address). The original patient registration number will be used as the unique patient identifier, rather than a project-specific unique identifier, to allow discussion of laboratory results and clinical process during the weekly project meeting (i.e. to prevent any miscommunication, mixing up of results). Original clinical and laboratory data forms will remain at the hospital laboratory and be archived according to local procedures. Copies will be stored in a locked filing cabinet at the reference laboratory. Data entered into databases will be password protected and stored on secure computers/servers. 12

23 Analysis of AST data Main objective (antibiotic susceptibility): the weighted average susceptibility for each of the main antibiotic groups used for empirical treatment of a BSI will be calculated. For the calculation of the weighted average, the total number of resistant bacteria is used as the numerator. The denominator is the total number of relevant microorganisms isolated. (See also Table 2, page 6); cumulative antibiograms (% sensitive/intermediate/resistant) for main antibiotic pathogen combinations will be produced according to standard procedures (24) and as described in detail in the CAESAR manual (1). Analysis may be stratified by patient characteristics; e.g. age, ward, source of infection (nosocomial or community-acquired). Secondary objectives (performance indicators): comparison of the results of species ID and AST performed in duplicate at local and central level, to ensure the quality of the data produced during the project; blood culturing practice - rate of blood culturing (number of BC sets per 1000 patient-days); - culture positivity rate (% BC with growth of pathogenic bacteria); - contamination rate (% BC with growth of a bacterium judged to be a contaminant); - proportion (%) of patients on antibiotics when BC is taken; timing of actionable laboratory results (median and interquartile range); - time to blood culture positivity (Gram stain result); - time to final blood culture results; proportions of adequate antibiotic treatment (percentages and 95% confidence intervals); - use of active and non-active therapy; - use of empirical therapy: unnecessary Gram-positive or Gram-negative coverage; and therapy change after laboratory result: de-escalation, escalation, expansion, optimization. Quality assurance Training Prior to initiation of the project, microbiologists from participating laboratories will be trained in blood culture procedures and techniques, species ID and EUCAST AST methodology. Training will consist of a theoretical and a practical part. Training will be facilitated by licensed physician(s) from one of the sponsoring institutes/organizations. At the start of the project, at each site, clinical staff will be trained (one or two days) in case identification and sampling procedures. In addition, physicians will be trained in principles of antibiotic treatment of 13

24 patients with sepsis and the incorporation of blood culture results into management to optimize antibiotic treatment (de-escalation, escalation, expansion of coverage). Feedback and continuous evaluation A weekly project meeting to discuss clinical sampling and laboratory results will be held at the participating hospital with clinicians, (clinical) microbiologists, epidemiologists, any additional staff involved in the project and the project coordinator; this will use the evaluation form (Annex 6). The quality of sampling and data collection, sampling rates, balance of patients from community/hospital, laboratory results, good practices, barriers and any problems/bottlenecks will be discussed. Additionally, the microbiologist will give feedback on diagnostic findings. Laboratory quality assurance Proficiency testing will be done throughout the project by testing all isolates in duplicate/parallel at local and central level. To ensure the quality of the data produced during the project and evaluate the impact of available blood culture results on clinical decision-making, the laboratory and clinical data forms will be reviewed by the infectious disease doctor/medical microbiologist of the technical support team at least once a month. In certain situations with unusual results (rare resistance phenotype), a subset of isolates can be tested at the facility of the infectious disease doctor/medical microbiologist of the technical support team. In addition, the laboratories will participate in the CAESAR external quality assessment, which will be organized via the WHO Regional Office for Europe. External quality assessment packages will be sent to the national reference laboratory for further distributed to the laboratories. Project management The project will be coordinated by the project coordination team. Full terms of reference are included in Annex 2. In short the national coordination team is responsible for; monitoring progress of the project developing local versions of the project protocol and SOPs providing project materials (protocol, SOPs, project forms) providing laboratory supplies data management data quality assurance evaluating and improving procedures during weekly project meetings at hospitals answering practical questions about project procedures continuing education The project coordinator will be supported at the technical level by the project initiators. The project coordinator will report back to the national AMR focal point on a monthly basis or whenever necessary to discuss the progress and any problems encountered. Safety considerations Taking samples for blood culture from patients is considered standard care for patients with suspected BSI or sepsis (12). The associated risks are the same as for any type of venipuncture. Risk will be mitigated by ensuring that sampling is done by trained medical staff. Sterile equipment and instructions on hygienic 14

25 practice during sample taking will be provided. Sampling will be done in the hospital setting with a physician on call if adverse events occur. The sampling procedures (venepuncture) may result in haematoma, swelling, tenderness and inflammation at the puncture site; persistent bleeding; vasovagal response; and in rare cases thrombosis of the vein from trauma or infection, which can result in thrombophlebitis. Staff in participating laboratories will be trained in blood culturing, species ID and state-of-the-art AST methods prior to initiation of the project and will be provided with quality controlled materials. In addition, blood culture isolates will also be tested at the national reference laboratory for quality assurance and confirmatory testing. Physicians will be trained in the principles of antibiotic treatment of patients with sepsis and in incorporating blood culture results to optimize antibiotic treatment (de-escalation, escalation, expansion of coverage). However, there is a risk that laboratory results may be misinterpreted (in particular false-negative results) and lead to incorrect clinical decisions, such as improper switching to a lower level of antibiotics. Training will be given by a licensed infectious disease doctor/medical microbiologist from the European Union at the start of the project. To ensure the quality of the data produced during the project and evaluate the impact of available blood culture results on clinical decision making, the laboratory and clinical data forms will be reviewed by the infectious disease doctor/medical microbiologist of the technical support team at least once a month. The infectious disease doctor/ medical microbiologist will also be available for trouble-shooting throughout the project. Follow-up Treatment and follow-up of patients providing blood samples for culture remains the responsibility of the treating physician. Results from bacteriological diagnostics will be provided to the treating physician by the local bacteriology laboratory (or the external laboratory providing microbiological diagnostic services) as soon as they become available. Adjusting treatment of the patient based on bacteriological results remains the responsibility of the treating physician. The process of communication and feedback of results between the microbiologist and clinician will be registered on the feedback form (Annex 5) and will be discussed during the weekly meeting. Importantly, parallel microbiological testing at the national AMR reference centre is primarily for the purpose of quality assurance and additional AST (confirmatory testing), as opposed to direct patient diagnostics. Ethics Taking blood samples from patients with suspected BSI for microbiological diagnosis (blood cultures, species ID and AST) is considered standard care required to inform treatment decisions (12) in many European countries, and according to expert opinion is considered best practice. If a patient with suspected BSI declines participation in the PoP project, the option of taking a blood culture for further AMR diagnosis will still be offered to this patient, in accordance with the SOPs in the hospital. This project provides technical and material support to allow the collection and processing of these samples. Patients are not subject to any additional blood sampling beyond that needed for microbiological diagnosis. Also, patients will not be treated with any experimental or unregistered drugs as part of this project. Treatment decisions remain at the discretion of treating clinician. Indeed, treatment decisions will be influenced by feedback of results from the microbiological work-up and, therefore, this need to be reliable. The adherence to quality standards and the assessment of the quality of the microbiological laboratory work is an integral part of this project, as described under quality assurance. After final confirmation at the national AMR reference laboratory, isolates will be stored at the repository for further characterization and molecular typing. 15